Alcohol is one of the most commonly abused drugs in the US, with 10% of the population affected by alcohol dependence at some point in their lives. An important newfocus is the development of new medications to help alcohol dependent patients reduce their alcohol use and promote abstinence. Potential targets for new medications are the multiple neurotransmitter systems of the brain reward systems including gamma- aminobutyric acid (GABA), glutamate, dopamine, cannabinoid and opioid peptides, which appear to be involved in the reinforcing effects of alcohol. Compulsive drinking can be conceptualized as a sequence of complex behaviors that terminate in alcohol consumption. Such complex sequences of behavior can be modeled in laboratory animals using chained schedules of reinforcement. The proposed studies will utilize a procedure in which initially neutral cues (lights and tones) are presented during 3 distinct components of a chained schedule of reinforcement. Each component is associated with an operant response requirement that is correlated with a distinct cue. Fulfilling the response requirement in the second component is necessary to progress to the third and final component. The primary reinforcer (alcohol or a preferred non- alcoholic beverage; Tang) will be available for self-administration only in the final component. This procedure allows the measurement of behaviors associated with alcohol anticipation, seeking, consumption and reinforcement within the same experimental session. The current proposal will evaluate the efficacyof newly proposed alcohol medications (e.g., topiramate, baclofen, 3-PBC and SR141716) in reducing alcohol seeking and consumption. In addition, current FDA-approved medications (naltrexone and acamprosate) will also be evaluated for comparison. The following hypotheses will be tested: 1) Test drugs will decrease alcohol self-administration behaviors and the amount of alcohol consumed, 2) Test drugs will decrease the motivation to gain access to alcohol, as measured by delaying onset of drinking and by reducing the maximum amount of work the subject will engage in to gain access to alcohol, 3) Decreases in self- administration and consumption produced by the test drugs will be greater for alcohol than for Tang, and 4) Test drugs will decrease the motivation to gain access to alcohol more than for Tang. These studies will provide important, previously unavailable, information on the differential behavioral efficacy of compounds that target receptor mechanisms believed to be influential in maintaining alcohol drinking and alcohol- seeking behaviors that are relevant to compulsive alcohol use and relapse in humans.